Articles
CHARACTERISATION OF PLANT RESIDUE QUALITY FOR PREDICTION OF DECOMPOSITION AND NITROGEN RELEASE IN AGRICULTURAL SOILS
Article number
700_4
Pages
57 – 62
Language
English
Abstract
The nitrogen (N) supply from crop residues and organic fertilisers must be integrated to N fertiliser recommendations as carefully as possible.
Thus, prediction of carbon (C) and N mineralisation patterns of plant residues is important for both agronomic and environmental purposes.
In this collaborative project of five Nordic countries, we tested the success of stepwise chemical digestion (SCD; Van Soest analysis), near infrared reflectance (NIR) spectroscopy and residue N concentration in the prediction of C and N mineralisation dynamics.
One of the major objectives was to develop low-cost NIR analyses as an alternative method of residue quality characterisation.
A total of 249 plant materials were collected and their NIR spectra were measured.
According to NIR analysis, 113 plant residues of widely differing qualities were selected and analysed for total N and subjected to SCD. These three methods were used to partition plant residue C and N into litter pools in a mechanistic, dynamic decomposition model and to predict parameters in a number of empirical functions to describe net C and N mineralisation dynamics of 76 different plant materials.
C mineralisation was predicted almost equally well by NIR and SCD (r2=0.910.93) but clearly better than by N concentration (r2=0.85). N mineralisation was better predicted by SCD fractions (r2=0.53) than by N concentration (r2=0.50) and NIR (r2=0.45). The decomposition model initialised from SCD, NIR or N concentration performed almost equally well (r2=0.690.76). According to these results, NIR spectra and total N concentration are cost-effective alternatives for prediction of plant residue decomposition.
These methods could be used for plant residue characterisation in N recommendations.
Thus, prediction of carbon (C) and N mineralisation patterns of plant residues is important for both agronomic and environmental purposes.
In this collaborative project of five Nordic countries, we tested the success of stepwise chemical digestion (SCD; Van Soest analysis), near infrared reflectance (NIR) spectroscopy and residue N concentration in the prediction of C and N mineralisation dynamics.
One of the major objectives was to develop low-cost NIR analyses as an alternative method of residue quality characterisation.
A total of 249 plant materials were collected and their NIR spectra were measured.
According to NIR analysis, 113 plant residues of widely differing qualities were selected and analysed for total N and subjected to SCD. These three methods were used to partition plant residue C and N into litter pools in a mechanistic, dynamic decomposition model and to predict parameters in a number of empirical functions to describe net C and N mineralisation dynamics of 76 different plant materials.
C mineralisation was predicted almost equally well by NIR and SCD (r2=0.910.93) but clearly better than by N concentration (r2=0.85). N mineralisation was better predicted by SCD fractions (r2=0.53) than by N concentration (r2=0.50) and NIR (r2=0.45). The decomposition model initialised from SCD, NIR or N concentration performed almost equally well (r2=0.690.76). According to these results, NIR spectra and total N concentration are cost-effective alternatives for prediction of plant residue decomposition.
These methods could be used for plant residue characterisation in N recommendations.
Publication
Authors
T. Salo, B. Stenberg, C. Lundström, L.S. Jensen, S. Bruun, A. Pedersen, T.A. Breland, T. Henriksen, A. Korsaeth, H. Palmason, J. Gudmundsson
Keywords
carbon, C/N ratio, mineralisation, near infrared reflectance spectroscopy, NIR, Van Soest fractionation
Online Articles (50)